The use of computers in boat design

Hoo boy, this way of thinking is so wrong. While you are correct to think that computers have changed naval architecture, they are only power tools, and like power tools thier main charm is the ability to do things faster...not better, just faster. GIGO is still the operative word, and you will be amazed at how fast they can screw up a thousand times in row.

Really, myself, and others who are of those 50ish years can remember the days before computers entered the design office. They have made some things easier, some things faster, but they have not replaced skill. I myself have run card decks overnight just to confirm what I could do by hand in 15 minutes. I still have xeroxed pipe flow nomographs on my desk just because thay give "better" answers than piping solvers. Don't even start me on the limitation of computer drafting programs and how slow they are to make an original "drawing". And in the last year I saw a 6 month long very expensive CFD exercise to comfirm what I said instantly by inspection...the people paying for it just did not have the experience to know what they were looking at.

And there is the real danger why experinced naval architects are lery of computers. There is always the very real problem of computer programs making design decisions for you, you have to be able to tell if the answer is the correct one. You need to trust the program, and I myself only trust those that I have coded and checked.

And for what it is worth, it is not modern science that made ships safer, but rather social and finacial pressure. Ships became safer when society, about 1850, no longer tolerated the large loss of life by marginal ship owners trying to eek the last shilling of profit. The insurance clubs rules, the tonnage laws, the COREGS and light laws, and the move to give finite life to wooden cargo ships to force the move to iron, all contributed to safer ships. Very few ships who's cause of lost was known were lost to design flaws. Most were lost to strandings, and collisions, fire, and the like which could be regulated.

Finaly, the comparison to a wooden sailing ship to a modern yacht is so poor as to reinforce my belief that you really don't understand the skill of the naval architects of the time compared to the modern ones. That "toilet bowl 30" is an overcanvased toy (I know, I sailed with Olsens on SF Bay), unsuitable for the conditions that a sailing ship was expected to operate in: Force 5 to 9. At the windspeed and sea you would be lying ahull, they, with wooden masts and hemp cordage and canvas, would be making 7-8 knots to weather.

 

What uses by someone with experience do you see for computers and commercial software in a design office today?

 

Jehardiman,

I actually agree with you. This thread was seperated from another, and the first post i made actually agrued that a CAD program is nothing more than a power tool. And that much like a power tool they can help a great worker make something faster, cheaper, easier, and better. But they are in no way a replacement for the knowledge and skill of a truly skilled operator.

 

I agree that computer is just a tool, but there is a line where better tool allows to make better things. One sample: classic manual calculations of stability do not allow to consider trim associated with inclinations (ok, one can make it but very time consuming and nobody did in reality). For small craft with asymmetry of abovewater part the difference between calculations with and without trim is 5...20% of GZ, to unsafe side. Today it is required to include trim in stability calculations - see ISO12217; this is possible due to better tools.

(Actually these graphs were made for Ukrainian Register of Shipping that does not allow use of 'unapproved' software for stability calculations and still requiring manual calculations. Just quick way to show them that error in manual method might cause overestimated stability and safety would suffer)

B1000 is 10m semi-planing powerboat; F716 is 7.16m sailboat with centreboard and internal ballast.

 

Alik, that's a good chart. Perhaps a brief discussion on the importance of including trim in stability calculations would be in order on the Stability forum. Some of the inexpensive software I've looked at either doesn't automatically include trim angles. I suspect some of the folks designing boats may not be aware of the effect it can have, particularly with wedge shaped boats..

 

Classic naval architecture does not cover this issue as well.

 

Actually, this is the use of computers I most take exception to. Stability and real-time loading calculations. There is a feeling that a brute force computer analysis is better than a human examination of root cause or monitoring. More than that, it allows a false feeling of comfort when shaving the stability to an over fine line. This is real case of the GIGO law. You can never actually know the displacement of any vessel; it will always have some relative accuracy. And, as in the case of the MV Rocknes where 18 died, often the program does not find the limiting case because it is not told to look for it (i.e. using the default case limitations), when it is immediately obvious to the trained observer that you need to investigate a specific case.

 

I believe that computer is not analysis tool, it is good calculator with wide range of possibilities. Analysis is done by expert operator.

 

Computers are tools - like hammers and screw drivers. Just because you have a good hammer it doesn't make you a good designer or a good engineer. Kind of trite, but true.

I think Gonzo's comment on ugly boats has a lot of merit. My thoughts on ugly boats: Just because you have the tools or just because you are a good engineer it doesn't make you a good designer. Engineering is a discipline - Design is a different discipline.

Design and engineering, though closely related, are two different subjects. Others on this forum have advised time and time again that to design boats you need to be an engineer. I would argue that a good designer NEEDS a good engineer and/or a good naval architect, but he doesn't need to BE an engineer. Virtually every large yacht around, and many small boats from large corporate entities have designers, naval architects, and engineering teams working on them. The notion that designers have to be engineers seems to come from smaller shops where one or two guys do it all, and often the work may be well engineered, but not necessarily well designed and it often shows in the boats.

OK off the soap box now - time for lunch.

 

You have to start from definitions: who is designer, stylist, naval architect, engineer. To me, 'designer' is type of job but naval architect is a qualification. To add more, naval architect is no just a trained person with calculator; look at word 'architect' - it is indication of creative component in this profession. Naval architect can be a designer if he is involved in design; can also be surveyor, researcher, etc.

 

Yep, I have seen many things designed that do not have a viable engineering solution (i.e. powerpoint engineering). Conversely, I have seen many things engineered that don't have good design (i.e. who put THAT fastener there). I like to use what we call "The Greybeards"...get 4-5 senior engineers from different disciplines together in a room and let them bash on each other until a design pops out. Not often the absolute best, but workable and buildable on time and in budget.

 

Actually design is solution of particular task, and aesthetics is only one component of design. There are others such as comfort, performance, safety, economics. Many people mistreat 'design' as only creating the appearance, but this is styling, not design. Design involves engineering component as well as aesthetic component, on equal basis. This is true for industrial design/boat design where object of design should be usable.

 

This is quite an interesting discussion isn't it? However, there are a few points that have been overlooked.

Firstly, computers are everywhere, yes, even in that cheap analogue watch you bought because it wasn't digital and therefore didn't have a computer. In your TV, microwave, everywhere. That's a good thing. They do simple, repetitive tasks all day long for very little power and don't complain.

Secondly, computers aren't about to disappear, so objecting on principle is pointless. You'll just get overtaken and left in the early 1940s.

Thirdly, computers will faithfully reproduce what the programmer and operator tell them to do. Just because you didn't like the answer, don't assume it's wrong, it's more likely to be the right answer to your question. You did understand the question, didn't you?

Fourthly, where computers really win is in the speed of calculation. The old methods were formed around hand-calcs, and compromises were made to make the process practical. Well, guess what, we don't have to compromise as much now, and the new mathematical methods that are emerging make use of that, to provide better accuracy, and greater speed.

It is perhaps obvious that from all this the people who stand to lose most are those who don't see the benefit, or those who don't want to change the way they work. I see that a lot, and it is obvious that those people are getting left behind (very quickly). The people who win are the young, eager guys (and girls) who pick up this new technology and make it do what they want to solve their problem.

Whatever tool you use in whatever walk of life, you need to understand how to use it. This may mean completely changing your understanding of the problem you're trying to solve. That shouldn't be scary, because that's what good engineers are able to do.

Tim B.

 

Actually, it is the other way around. The calculations made by computers have compromises in order to be workable. There are many engineering problems that cannot be solved by a digital machine no matter how many FPO's you throw at it. It is the nature of the beast, just like CAD programs cannot draw an correct involute sprial. The question is wether the accuracy is acceptable...in some cases it is not.

 

I was about to say something similar, but I was going to use "turbulence" and "3D planing" as examples. The involute spiral is a terrific simple example!


.